JP2010250657A - Printing apparatus, image processing apparatus, image processing method and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve user convenience in image processing that performs an image search. <P>SOLUTION: An image processing apparatus includes: a condition designation window display unit that displays a condition designation window which includes condition designation regions by stages for designating search conditions for features of image contents in a plurality of search stages in series relations with one another, and a plurality of tags that designate display states of the condition designation regions by stages as tags corresponding to the plurality of search stages; a search condition setting unit that sets the search conditions in the plurality of search stages in accordance with the designations through the condition designation window; and an image search unit that sequentially performs image searches for the plurality of search stages by using the set search conditions. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to image processing for performing image search.

  A search condition related to image attributes (for example, shooting date and time and shooting mode) and a search condition related to image content characteristics (for example, similarity to a predetermined template image) are set, and an image that satisfies the search condition from a plurality of images Image processing for performing image search for detecting the image is known (see, for example, Patent Document 1).

JP 2004-272314 A

  In the conventional image processing for performing image search, there is room for improvement in terms of user convenience when specifying a search condition used for image search.

  SUMMARY An advantage of some aspects of the invention is to improve user convenience in image processing for performing an image search.

  In order to solve at least a part of the above problems, the present invention can be realized as the following forms or application examples.

Application Example 1 An image processing apparatus that performs an image search,
A stage-specific condition designation area for designating a search condition relating to the characteristics of image contents in each of a plurality of search stages having a serial relationship, and a tag corresponding to each of the plurality of search stages, the stage-specific condition designation area A condition specification window display for displaying a condition specification window including a plurality of tags for specifying the display state of
A search condition setting unit that sets the search condition in each of the plurality of search stages according to the specification through the condition specifying window;
An image search unit that sequentially executes image search of the plurality of search stages using the set search conditions.

  In this image processing apparatus, a condition designation window is displayed, search conditions in each of a plurality of search stages are set according to the designation through the condition designation window, and image retrieval of a plurality of search stages using the set search conditions is performed. It is executed sequentially. Here, the condition designation window is a condition designation area for each stage for designating a search condition related to the feature of the image content in each of a plurality of search stages having a serial relationship, and tags corresponding to each of the plurality of search stages. And a plurality of tags for designating the display state of the stage-specific condition designation area. Therefore, the user can specify the search condition regarding the feature of the image content in each of the plurality of search stages while switching the display state of the stage-specific condition specifying area by the tag. Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

[Application Example 2] The image processing apparatus according to Application Example 1,
The stage-specific condition designation area is an area for designating one of the plurality of search conditions that realizes an image search in which at least one of processing speed and processing accuracy is different as the search condition to be adopted. Processing equipment.

  In this image processing apparatus, the user adopts one of a plurality of search conditions that realize image search in which at least one of processing speed and processing accuracy is different from each other via the stage-specific condition specifying area of the condition specifying window. The search condition can be specified as a search condition to be performed, and a search condition for realizing an image search with a desired processing speed and processing accuracy can be easily set. Therefore, this image processing apparatus can improve user convenience in image processing.

[Application Example 3] The image processing apparatus according to Application Example 1 or Application Example 2,
The condition designating window display unit displays the condition designating window including a stage number designating area for designating the number of the search stages,
The image processing apparatus, wherein the search condition setting unit sets the search condition in each of a number of the search stages specified via the condition specifying window.

  In this image processing apparatus, a condition designation window including a stage number designation area for designating the number of search stages is displayed, and search conditions for each of the number of search stages designated via the condition designation window are set. The user can easily specify the number of search stages (search stages) in which search conditions can be set independently. Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

Application Example 4 The image processing apparatus according to any one of Application Examples 1 to 3, further comprising:
A query image setting unit for setting a query image as a reference for image search;
The said search condition setting part is an image processing apparatus which sets the said search condition regarding the similarity with the said query image about the said characteristic.

  In this image processing apparatus, a query image as a reference for image search is set, a search condition regarding similarity to the query image with respect to the feature of the image content is set, and a plurality of search stages using the set search condition are set. Image retrieval is performed sequentially. Therefore, in this image processing apparatus, user convenience can be improved in image processing in which an image search is performed using a search condition relating to the similarity to the query image with respect to the feature of the image content.

[Application Example 5] The image processing apparatus according to Application Example 4,
The query image setting unit sets an image specified by any one of an image file designation, a drawing, and a color designation as the query image.

  In this image processing apparatus, an image specified by any one of image file designation, drawing, and color designation is set as a query image, so that a user can easily select a desired image as a query image. Can be set to Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

Application Example 6 The image processing apparatus according to Application Example 4 or Application Example 5,
The said query image setting part is an image processing apparatus which sets one of the images detected by the image search as a new said query image.

  In this image processing apparatus, since one of the images detected by the image search is set as a new query image, for example, the user is more likely to set a detected image that is considered closer to the desired image as a new query image. It becomes possible to reach a desired search result efficiently. Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

Application Example 7 The image processing apparatus according to any one of Application Example 4 to Application Example 6,
The said search condition setting part is an image processing apparatus which sets the said search condition regarding the said similarity weighted for every area | region of the image.

  In this image processing apparatus, it is possible to set a search condition related to the similarity weighted for each area of the image, and the user can perform an image search that emphasizes (or neglects) a specific area of the image, A desired image can be searched more easily and efficiently. Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

Application Example 8 The image processing apparatus according to any one of Application Example 4 to Application Example 7,
The image processing apparatus, wherein the search condition setting unit sets the search condition related to the similarity weighted for each channel of an image color space.

  In this image processing apparatus, it is possible to set a search condition related to the degree of similarity weighted for each channel in the color space of the image, and the user performs an image search that emphasizes (or neglects) a specific channel of the image. The desired image can be searched more easily and efficiently. Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

[Application Example 9] The image processing apparatus according to any one of Application Example 1 to Application Example 8,
The image search unit is an image processing device that performs an image search of the subsequent search stage for an image detected in the image search of the previous search stage in the serial relationship.

  In this image processing apparatus, since the image search of the subsequent search stage is performed on the image detected in the image search of the previous search stage in the serial relation, the subsequent search stage is adjusted by adjusting the search conditions of the previous search stage. Since the number of target images can be adjusted, user convenience can be improved in image processing for performing image search.

Application Example 10 The image processing apparatus according to any one of Application Example 1 to Application Example 9,
The search condition setting unit sets a search condition related to the attribute of the image in addition to the search condition related to the feature of the image content,
The image search unit is an image processing device that performs an image search using a search condition relating to an attribute of the image and a search condition relating to a feature of the image content.

  In this image processing apparatus, in addition to the search condition relating to the feature of the image content, the search condition relating to the attribute of the image is set, and the image search using the search condition relating to the attribute of the image and the search condition relating to the feature of the image content is performed. Therefore, an image search for detecting a desired image can be realized more easily and efficiently, and user convenience can be improved in image processing for performing an image search.

[Application Example 11] The image processing apparatus according to any one of Application Example 1 to Application Example 10,
The condition designation window display unit displays the condition designation window in which the stage-specific condition designation areas corresponding to only one search stage are displayed at the same time according to designation via the tag.

  In this image processing apparatus, since the condition specifying area for each stage corresponding to only one search stage is displayed at the same time in the condition specifying window, the user can specify the condition in the displayed condition specifying area for each stage, A search condition for one search stage can be easily and reliably specified. Therefore, in this image processing apparatus, user convenience can be improved in image processing for performing image search.

Application Example 12 The image processing apparatus according to any one of Application Example 1 to Application Example 11,
The image processing apparatus, wherein the feature is at least one of a feature representing a color distribution in an image and a feature calculated by wavelet decomposition of the image.

  In this image processing apparatus, a user in image processing that performs an image search specifying a search condition for at least one of a feature representing color distribution in an image and a feature calculated by wavelet decomposition of the image as a feature of the image content. Convenience can be improved.

Application Example 13 The image processing apparatus according to any one of Application Examples 1 to 12, further comprising:
A print image setting unit for setting a detection image to be printed from detection images detected by image search;
An image processing apparatus comprising: a printing unit that prints a detection image to be printed.

  In this image processing apparatus, the detection image to be printed is set from the detection images detected by the image search, and the detection image to be printed is printed. Therefore, the user is required to print the image detected by the image search. Convenience can be improved.

[Application Example 14] The image processing apparatus according to Application Example 13, further comprising:
A search result window display unit for displaying a search result window including a detection image display area for displaying a list of detection images and a print designation area for designating a detection image to be printed;
The print image setting unit, in the search result window, sets a detection image corresponding to an image displayed in the print designation area as a detection image to be printed in response to a user's predetermined operation.

  In this image processing apparatus, a search result window including a detection image display area for displaying a list of detection images and a print designation area for designating detection images to be printed is displayed. In the search result window, a predetermined operation by a user is displayed. Accordingly, since the detection image corresponding to the image displayed in the print designation area is set as the detection image to be printed, the user can easily designate the detection image to be printed.

  It should be noted that the present invention can be realized in various modes. For example, an image processing method and apparatus, an image search method and apparatus, a printing method and apparatus, and a computer for realizing the functions of these methods or apparatuses The present invention can be realized in the form of a program, a recording medium recording the computer program, a data signal including the computer program and embodied in a carrier wave, and the like.

1 is an explanatory diagram schematically showing a configuration of a printer 100 as an image processing apparatus in a first embodiment of the present invention. FIG. 6 is a flowchart showing a flow of image search / print processing in the first embodiment. It is explanatory drawing which shows an example of the initial window W1. It is explanatory drawing which shows an example of the initial window W1. It is explanatory drawing which shows an example of search option window W2. It is explanatory drawing which shows an example of search option window W2. It is explanatory drawing which shows an example of the search condition choice of metadata utilization search. It is explanatory drawing which shows an example of the search condition choice of an image content use search. It is explanatory drawing which shows an example of search option window W2. It is a flowchart which shows the flow of a search execution process. It is explanatory drawing which shows an example of search result window W3. It is explanatory drawing which shows an example of search option window W2. It is explanatory drawing which shows roughly the structure of the printer 100a as an image processing apparatus in 2nd Example. It is a flowchart which shows the flow of the image search and printing process in 2nd Example. It is a flowchart which shows the flow of a search condition setting process. It is explanatory drawing which shows an example of choice table CT. It is explanatory drawing which shows a mode that a search stage and a search condition are set in a search condition setting process.

Next, embodiments of the present invention will be described in the following order based on examples.
A. First embodiment:
A-1. Configuration of image processing device:
A-2. Image search / print processing:
A-3. Feature distance calculation method:
A-3-1. Regarding the color histogram:
A-3-2. Regarding Haar wavelets:
B. Second embodiment:
C. Variation:

A. First embodiment:
A-1. Configuration of image processing device:
FIG. 1 is an explanatory diagram schematically showing the configuration of a printer 100 as an image processing apparatus according to the first embodiment of the present invention. The printer 100 of this embodiment is an ink jet color printer that supports so-called direct printing, in which an image is printed based on image data acquired from a memory card MC or the like. The printer 100 includes a CPU 110 that controls each unit of the printer 100, an internal memory 120 configured by a ROM and a RAM, an operation unit 140 configured by buttons and a touch panel, a display unit 150 configured by a liquid crystal display, and a printer. An engine 160 and a card interface (card I / F) 170 are provided. The printer 100 may further include an interface for performing data communication with other devices (for example, a digital still camera or a personal computer). Each component of the printer 100 is connected to each other via a bus.

  The printer engine 160 is a printing mechanism that performs printing based on print data. The card interface 170 is an interface for exchanging data with the memory card MC inserted into the card slot 172. In this embodiment, an image file including image data is stored in the memory card MC.

  The internal memory 120 stores an image processing unit 200, a display processing unit 310, and a print processing unit 320. The image processing unit 200 is a computer program for executing image search / print processing under a predetermined operating system. The image search / printing process according to the present embodiment is a process of performing an image search for detecting an image that matches a search condition and printing the detected image. The image search / print process will be described in detail later.

  The image processing unit 200 includes an image search unit 210 as a program module. The image search unit 210 includes a window display control unit 211, a query image setting unit 212, a search condition setting unit 213, a similarity calculation unit 216, and a print image setting unit 217. The functions of these units will be described in detail in the description of image search / print processing described later.

  The display processing unit 310 is a display driver that controls the display unit 150 to display a window screen including a processing menu and a message, an image, and the like on the display unit 150. The print processing unit 320 is a computer program for generating print data from image data, controlling the printer engine 160, and printing an image based on the print data. The CPU 110 reads out and executes these programs (the image processing unit 200, the display processing unit 310, and the print processing unit 320) from the internal memory 120, thereby realizing the functions of these units.

A-2. Image search / print processing:
FIG. 2 is a flowchart showing the flow of image search / print processing in the first embodiment. The image search / printing process of the present embodiment is a process of performing an image search for detecting an image that meets a search condition from a plurality of target images, and printing the detected image. The printer 100 according to the present exemplary embodiment can execute an image search based on the feature of the image content in addition to the image search based on the image attribute. The image search based on the feature of the image content is an image having a high degree of similarity (similarity) with the query image as a reference in the image search with respect to a predetermined feature of the image content from the target images to be searched. This is done by detecting. More specifically, a feature amount distance representing the similarity between the query image related to a predetermined feature of the image content and the target image is calculated, and a target image whose feature amount distance is equal to or less than the maximum feature amount distance is detected. Although the target image can be arbitrarily set, in this embodiment, it is assumed that an image stored in the memory card MC (image data included in the image file) is set as the target image.

  When image search / print processing (FIG. 2) is started in response to an instruction from the user via the operation unit 140 (FIG. 1), the window display control unit 211 controls the display processing unit 310 to display the display unit. The initial window W1 is displayed at 150 (step S110). FIG. 3 is an explanatory diagram showing an example of the initial window W1. The initial window W1 is a window including a query image for image search and a user interface for designating an image search type.

  In this embodiment, it is possible to specify a query image by three different methods. That is, as a method for specifying a query image, a method (T1) by designating an image file, a method (T2) by drawing, and a method (T3) by color selection are prepared. As shown in FIG. 3, the initial window W1 includes user interfaces corresponding to the three query image designating methods.

  That is, in the initial window W1 (FIG. 3), as an interface for the method (T1) by designating an image file, an input box Bo11 for inputting a path for specifying the position of the image file and a hierarchical selection of the image file are selected. And a browse button Bu13 for displaying a window for the purpose. The user designates an image to be set as a query image by directly inputting the path of the image file in the input box Bo11 or by selecting an image file in a window displayed by pressing the browse button Bu13. be able to.

  In addition, in the initial window W1 (FIG. 3), as an interface for the drawing method (T2), a color palette button Bu23 including a plurality of color buttons of different colors and a stroke diameter specification including a plurality of brush buttons of different diameters. And a button Bu24. The user can designate an image to be set as a query image by drawing an image in the drawing area Ar21 using the color palette button Bu23 or the stroke diameter designation button Bu24. The initial window W1 also includes a clear button Bu25 for erasing the image. The drawing process in the initial window W1 is realized by using, for example, a Java script (registered trademark) drawing tool.

  The initial window W1 (FIG. 3) is an input for inputting the value of each channel in a predetermined color system (for example, HSV color system or RGB color system) as an interface for the color selection method (T3). Box Bo31, color designation area Ar31 for selecting a color by designating one point in the color gradation, and a position along a specific color axis (for example, a hue axis or a saturation axis) The slider Sl31 is included. The user inputs a value of each channel in a predetermined color system in the input box Bo31, designates one point in the color designation area Ar31, or designates a position along the axis in the slider Sl31. One specific color can be selected, whereby an image to be set as a query image (a solid image composed of the selected one color) can be designated. The color selected by the user is displayed in the color display area Ar32. In the input box Bo31, the values of the channels in the plurality of color systems are associated with each other. For example, when the value of a channel in the HSV color system is changed, the values in the RGB color system and the CMY color system are used. The value of each channel is automatically changed. The radio button in the input box Bo31 is for designating the axis in the slider Sl31. For example, when the radio button at the position of the S channel in the HSV color system is selected, the slider Sl31 is set to the saturation axis. It will be corresponding. Further, a plurality of interfaces for the method (T3) by color selection in the initial window W1 are also associated with each other. For example, when one point in the color designation area Ar31 is designated, each table representing the color of the designated point A value in the color system is displayed in the input box Bo31. In the code display box Bo32, a code representing the designated color is displayed. The color selection process in the initial window W1 is realized using, for example, a Java script color palette.

  In this embodiment, two types of image search types, a normal search and a quick search, are prepared. Normal search is a search type in which search conditions are set in detail and image search is executed. Quick search does not perform detailed settings of search conditions, and performs image search using the default search conditions. Search type. In the initial window W1, for each of the above three methods (T1, T2, T3) for setting a query image, a button (Bu11, Bu21, Bu31) for starting a normal search and a quick search are started. Button (Bu12, Bu22, Bu32).

  When one of the buttons for starting the image search is selected by the user in the initial window W1 (FIG. 3), the query image setting unit 212 (FIG. 1) sets the query image (step S120 in FIG. 2), The image search unit 210 sets a search type (step S130). That is, the search type (normal search or quick search) corresponding to the button selected by the user is set as the search type to be used, and the image specified by the query image specifying method corresponding to the selected button is displayed. Are set as query images. FIG. 4 shows an example of the initial window W1 when a query image is designated by drawing. For example, as shown in FIG. 4, when the user draws an image in the drawing area Ar21 and selects the button Bu21 for starting a normal search corresponding to the query image designation method by drawing, the image generated by the drawing Is set as the query image, and normal search is set as the search type.

  When the button (Bu11, Bu21, Bu31) for starting a normal search is selected in the initial window W1 (FIG. 3) and the normal search is set as a search type (step S140: No in FIG. 2), the window The display control unit 211 (FIG. 1) controls the display processing unit 310 to display the search option window W2 on the display unit 150 (step S150). FIG. 5 is an explanatory diagram showing an example of the search option window W2. The search option window W2 is a window including a user interface for specifying search conditions in image search in detail. The search option window W2 corresponds to a condition designation window in the present invention, and the window display control unit 211, the display processing unit 310, and the display unit 150 function as a condition designation window display unit in the present invention.

  As shown in FIG. 5, the search option window W2 includes a query image display area Ar41 for displaying a query image, and an attribute display area Ar42 for displaying attributes (file name and file size) of the query image. Yes. Further, as described above, the printer 100 according to the present exemplary embodiment can perform an image search based on metadata representing image attributes (hereinafter also referred to as “metadata use search”) and an image search based on image content characteristics (hereinafter referred to as “metadata use search”). , Also called “image content use search”). Therefore, the search option window W2 includes a metadata condition designation area Ar43 for designating search conditions for metadata usage search, and a content condition designation area Ar45 for designating search conditions for image content usage search. Yes.

  In the search option window W2 (FIG. 5), a plus button Bu45 arranged near the metadata condition specifying area Ar43 is a button for adding a search condition for metadata use search. When the plus button Bu45 is selected, as shown in FIG. 6, a set of boxes for specifying one search condition for metadata use search is additionally displayed in the metadata condition specification area Ar43. The minus button Bu46 is a button for deleting one search condition for the metadata use search.

  As shown in FIG. 6, a set of boxes for specifying one search condition for metadata use search includes a box Bo51 for specifying a metadata type and a box Bo52 for specifying a metadata item. And a box Bo53 for designating an inequality sign in the search condition and a box Bo54 for designating a value in the search condition. The designation in the boxes Bo51, 52, 53 is performed by selection from the pull-down menu. FIG. 7 is an explanatory diagram showing an example of search condition options for the metadata usage search. In the present embodiment, as shown in FIG. 7, two types of information, ie, Exif information and general information, are set as metadata type options (options in the box Bo51). In addition, as options corresponding to Exif information, camera manufacturer, camera model, shooting date and time, etc. are set as metadata item options (options in the box Bo52). A size, an image size (width, height), the number of detected face images, and the like are set. The user can specify a search condition such as “file size is smaller than 1,000 kB” by selecting a desired option in the boxes Bo51, 52 and 53 and inputting a value in the box Bo54.

  When a plurality of metadata use search conditions are specified in the metadata condition specification area Ar43, a box for specifying whether the relationship between the search conditions is “and” or “or” (not shown). ) Is displayed, and the relationship between the search conditions can be specified. Further, if no search condition is specified in the metadata condition specifying area Ar43, the metadata use search is not performed.

  The printer 100 according to the present exemplary embodiment can set a plurality of search stages (search stages) having a serial relationship in an image content use search. When a plurality of search stages are set in the image content use search, an image detected as satisfying the search condition in the previous search stage in the serial relationship becomes the target image in the subsequent search stage.

  In the search option window W2 (FIG. 5), a plus button Bu43 arranged near the content condition designation area Ar45 is a button for adding a search stage for image content use search. When the plus button Bu43 is selected, one search stage is added, and a search stage defining area Ar46 that defines search conditions in one search stage is added to the content condition specifying area Ar45. The minus button Bu44 is a button for deleting one search stage. Each search stage is associated with a tag Ta41 for designating the display state of the search stage definition area Ar46. When a plurality of search stages are set, one search stage is selected according to the selection of the tag Ta41. The corresponding search stage definition area Ar46 is displayed, and the search stage definition area Ar46 corresponding to another search stage is not displayed. In the search option window W2, the plus button Bu43 and the minus button Bu44 correspond to the stage number designation area in the present invention, and the search stage definition area Ar46 corresponds to the stage-specific condition designation area in the present invention. If no search stage is set in the content condition designation area Ar45, the image content use search is not performed.

  The search stage definition area Ar46 in the search option window W2 (FIG. 5) is an area for designating search conditions in the corresponding search stage. In the present embodiment, when performing an image search using a search condition, it is possible to specify the search condition by selecting from a plurality of search condition options in which at least one of processing speed and accuracy is different from each other. It has become. The search stage defining area Ar46 includes a condition specifying area Ar47 for specifying a search condition and a weight specifying area Ar48 for specifying a weight for each area obtained by dividing an image into a plurality of areas. As shown in FIG. 5, the condition specifying area Ar47 includes a box Bo41 for specifying a signature method, a box Bo42 for specifying a metric type, a box Bo43 for specifying a color space, and each color space. A box Bo44 for specifying the weight of the color channel, a box Bo45 for specifying the maximum feature amount distance, and a box Bo46 for specifying the maximum number of detections are included.

  The designation in the boxes Bo41, 42, 43 in the condition designation area Ar47 (FIG. 5) is performed by selection from the pull-down menu. FIG. 8 is an explanatory diagram illustrating an example of search condition options for the image content use search. The color space specified by the box Bo43 is a color space used for setting search conditions. In this embodiment, RGB, LAB, YUV, and HSV are set as color space options as shown in FIG. Has been.

  The signature method designated by the box Bo41 in the condition designation area Ar47 (FIG. 5) is a feature of the image content used for the image content use search. In this embodiment, as shown in FIG. 8, two options of a color histogram and a Haar wavelet are set as signature method options. The metric type specified by the box Bo42 is a method for calculating the similarity between the query image and the target image related to a predetermined feature of the image content (an index value indicating the similarity and a method for calculating the index value). . In this embodiment, as shown in FIG. 8, standard histograms, correlation histograms, color moments, and combined features are set as metric type options corresponding to color histograms, which correspond to Haar wavelets. As a metric type option, a high-speed and low-precision metric M1, a medium-speed and medium-precision metric M2, a low-speed and high-precision metric M3, and a metric M4 as a real metric are set.

  A color histogram (see FIG. 8) as a signature method is a feature representing the color distribution of an image expressed in a predetermined color space. In this embodiment, the pixel frequency distribution in each of a plurality of color bins set by quantizing each of the channels defining the color space is used as the color histogram. For example, when an RGB color space is used, each RGB channel is quantized into four equal parts (that is, when the value range of each channel is 0 to 255, for example, 0 to 63, 64 to 127). , 128 to 191, and 192 to 255, which are quantized into four ranges), the pixel frequency distribution in each of the 64 color bins set is used. When the color histogram is used as the signature method, the image search speed is improved because the amount of calculation is small compared to the case where the Haar wavelet is used, while the color histogram has no spatial information. Often the accuracy of the search is reduced. For example, even if the image has a low similarity in terms of human interpretation, there is a possibility that an image having similar color distribution characteristics as the entire image may be detected. Note that the level of quantization when setting color bins need not be the same for each channel. For example, when the HSV color space is used, the H channel may be quantized more finely than the S channel and the V channel.

  The standard histogram (see FIG. 8) as a metric type option corresponding to the color histogram uses the pixel frequency itself in each of the plurality of color bins to calculate the similarity between the query image related to the color histogram and the target image. Is the method. The correlation histogram is a method for calculating the similarity between the query image and the target image using the cumulative value of the pixel frequency in the color bin. For example, assuming that the pixel frequency in each of the N color bins is Hi (i = 1, 2, 3,..., N), the cumulative value Cj (j = 1, 2, 3,. N) is calculated by the following equation (1). When the correlation histogram is adopted as the metric type, the image search speed is slightly lower than that of the standard histogram, but the occurrence of mismatch due to quantization error can be suppressed. Accuracy is improved.

Further, the color moment (see FIG. 8) as a metric type option corresponding to the color histogram uses three color moments of an average μ of pixel frequencies, a variance σ ² , and a skewness γ ³ in a color bin. This is a method of calculating the similarity between the query image related to the histogram and the target image. The combined feature is a method of calculating the similarity between the query image and the target image using an index F obtained by combining ^three color moments (average μ, variance σ ² , and skewness γ ³ ). The index F combined with the average μ, the variance σ ^2, and the skewness γ ³ is calculated by the following equations (2) to (5), respectively. Incidentally, a weighting coefficient is W _F in equation (5) is set by experiment. When a color moment or combined feature is adopted as the metric type, the image search speed is faster than the standard histogram, but the image search accuracy is lowered.

  The Haar wavelet as a signature method is a wavelet coefficient calculated by wavelet decomposition of an image using the Haar wavelet as a base. When the Haar wavelet is used as the signature method, the image search speed is reduced due to a large amount of calculation compared to the case where the color histogram is adopted, whereas the Haar wavelet has the spatial information of the image, so In many cases, the accuracy of search is improved.

  The high-speed low-accuracy metric M1, the medium-speed medium-accuracy metric M2, and the low-speed high-accuracy metric M3, which are metric type options corresponding to the Haar wavelet, all have Haar wavelet coefficients calculated by Haar wavelet decomposition of the image. This is a method of calculating the similarity between the query image and the target image. The high-speed low-accuracy metric M1, the medium-speed medium-accuracy metric M2, and the low-speed high-accuracy metric M3 have different image resolutions when performing Haar wavelet decomposition. That is, the high-speed and low-precision metric M1 is a method that uses the Haar wavelet coefficient calculated by Haar wavelet decomposition for a relatively low-resolution image, and the medium-speed and medium-precision metric M2 has a medium resolution. This is a method of using Haar wavelet coefficients calculated by Haar wavelet decomposition for an image of the same, and the low-speed and high-precision metric M3 is a Haar wavelet calculated by Haar wavelet decomposition for a relatively high-resolution image. This is a method using a coefficient. The higher the resolution of the image to be subjected to Haar wavelet decomposition, the slower the image search speed, but the more accurate the image search.

  The real metric M4 as a metric type option is also a method of calculating the similarity between the query image and the target image by using the Haar wavelet coefficient calculated by Haar wavelet decomposition of the image. The real metric M4 is different from the metric M1, the metric M2, and the metric M3 in that a symmetric method is used when calculating the similarity (feature amount distance) between the query image and the target image.

  A specific method for calculating the similarity (feature amount distance) between the query image and the target image relating to a predetermined feature of the image will be described later in “A-3. Feature amount distance calculation method”.

  When Haar wavelet is selected as the signature method, a box Bo49 for designating the number of wavelet coefficients Cu is displayed in the condition designating area Ar47 of the search option window W2, as shown in FIG. The wavelet coefficient number Cu is the number of wavelet coefficients (per channel) used when calculating the similarity (feature amount distance) between the query image and the target image related to the Haar wavelet. It can be said that the wavelet coefficient number Cu is an index value representing a small image compression rate. That is, assuming that the total number of Haar wavelet coefficients is Ca, the (Ca-Cu) wavelet coefficients having the smallest value are discarded (that is, the value is made zero). The smaller the wavelet coefficient number Cu, the higher the speed and accuracy of the similarity calculation process, and the higher the wavelet coefficient number, the lower the speed and precision of the similarity calculation process.

  The weight of each color channel in the color space designated by the box Bo44 in the condition designation area Ar47 (FIG. 5) is the weight of each color channel reflected in the calculation of the similarity between the query image and the target image regarding the feature of the image content. It is. The maximum feature amount distance designated by the box Bo45 in the condition designation area Ar47 is a threshold value of the similarity (feature amount distance) between the query image and the target image in the image search. That is, at the search stage, a target image whose similarity (feature amount distance) between the query image and the target image is equal to or less than the maximum feature amount distance is detected. The maximum number of detections designated by the box Bo46 in the condition designation area Ar47 is the allowable maximum number of target images that are detected as meeting the search condition in the search stage.

  The weight specification area Ar48 of the search stage definition area Ar46 in the search option window W2 (FIG. 5) is divided into a plurality of (16 in this embodiment) areas in a grid pattern, and each area has a weight specification box Bo47. A check box Bo48 is arranged. A query image is displayed in the weight designation area Ar48, and thereby the query image is divided into a plurality of regions. The weight designation box Bo47 is a box for designating the weight of each area of the image in calculating the similarity (feature amount distance) between the query image and the target image. In addition, by deselecting the check box Bo48 of a certain divided area, it is possible to specify that the area is not considered at all in calculating the similarity between the query image and the target image.

  The user selects a signature method, a metric type, a color space, and a weight for each channel in the condition specification area Ar47 and the weight specification area Ar48 of the search stage definition area Ar46 in the search option window W2 (FIG. 5). Or the weight of each region of the image can be specified, and the method of calculating the similarity (feature amount distance) between the query image and the target image at each search stage can be individually specified. In addition, the user can designate the maximum feature amount distance in the condition designation area Ar47 to designate the threshold value of each search stage, that is, how much a target image similar to the query image is detected in each search stage. it can. In addition, the user can designate the upper limit of images detected in each search stage by designating the maximum number of detections in the condition designation area Ar47.

  When the user specifies a search condition in the metadata condition specification area Ar43 of the search option window W2 (FIG. 5), the search condition setting unit 213 (FIG. 1) sets the search condition in the metadata use search (FIG. 1). 2 step S160). When the number of search stages is specified in the content condition specification area Ar45 of the search option window W2 by the user, the search condition setting unit 213 sets the number of search stages for image content use search (step S170). When the search condition is specified in the search stage definition area Ar46 by the user, the search condition setting unit 213 sets the search condition in each search stage of the image content use search (step S180). The search condition setting unit 213 also sets the execution order of each search stage based on the designation in the search stage definition area Ar46. In this embodiment, a stage number (such as “stage 1” in FIG. 5) is assigned to each search stage, and the search stages are executed in ascending order.

  The search option window W2 (FIG. 5) includes a search execution button Bu41 and a return button Bu42. When the return button Bu42 is selected in the search option window W2, the initial window W1 (FIG. 3) is displayed again on the display unit 150 (FIG. 1). When the search execution button Bu41 is selected in the search option window W2, search execution processing (step S190 in FIG. 2) is started.

  When the user selects a button (Bu12, Bu22, Bu32) for starting a quick search in the initial window W1 (FIG. 3) (step S140: Yes in FIG. 2), the search option window W2 is used. The detailed search condition setting (steps S150 to S180) is skipped, and the search execution process (step S190) using the default search condition is started.

  FIG. 10 is a flowchart showing the flow of search execution processing. In step S410, the image search unit 210 (FIG. 1) performs condition determination based on the metadata. Specifically, the image search unit 210 analyzes the image file of the target image to acquire metadata, and the attribute of the target image specified by the acquired metadata is a search condition regarding the set metadata ( For example, it is determined whether or not a search condition such as “file size is smaller than 1,000 kB” (see FIG. 6) is satisfied. The image search unit 210 excludes the target image determined not to meet the search condition from the subsequent search processing targets. If one search condition is not specified in the metadata condition specification area Ar43 of the search option window W2 (FIG. 5), the process of step S410 is skipped.

  In step S420 (FIG. 10), the image search unit 210 (FIG. 1) selects one of the search stages for the image content use search. As described above, in this embodiment, the search stages are selected in order from the search stage having the smallest stage number. For example, as shown in FIG. 5, when two search stages of a search stage “stage 1” of stage number 1 and a search stage “stage 2” of stage number 2 are set, first, the search stage “stage” 1 "is selected.

  In step S430 (FIG. 10), the image search unit 210 (FIG. 1) performs condition determination based on the feature amount distance. Specifically, the similarity calculation unit 216 of the image search unit 210 calculates the similarity (feature amount distance) between the query image and the target image relating to the feature of the image content according to the search condition set for the selected search stage. The face feature position specifying unit 210 calculates whether or not the calculated feature amount distance is equal to or less than the maximum feature amount distance. The image search unit 210 excludes the target image determined that the calculated feature amount distance is larger than the maximum feature amount distance from subsequent search processing targets.

For example, in the search conditions set for the selected search stage, as shown in FIG. 5, the signature method is “color histogram”, the metric type is “standard histogram”, and the color space is “RGB”. In some cases, the pixel frequency H _i of each color bin in the RGB color space is calculated as a feature value for each of the query image and the target image, and the feature value represents the similarity between the feature values (pixel frequency H _i ) of both. A distance is calculated. Then, it is determined whether or not the calculated feature amount distance is equal to or less than a maximum feature amount distance (for example, 100). When the signature method is “Haar wavelet”, the metric type is “metric M1”, and the color space is “RGB”, the Haar wavelet coefficient in the RGB color space for each of the query image and the target image. Is calculated, and a feature distance indicating the similarity between the two is calculated based on the Haar wavelet coefficient. When the weight of the color channel is set in the box Bo44 of the condition specifying area Ar47 or when the weight of the divided area is set in the weight specifying area Ar48, the feature amount distance is calculated in consideration of the weight. Is done. Further, the feature amount of the query image needs to be calculated only once. In addition, the feature amount of the target image or query image may be calculated in advance before the search execution process. The same applies to other signature methods and metric types. A method for calculating the feature distance will be described in detail later.

  In step S440 (FIG. 10), the image search unit 210 (FIG. 1) determines whether or not the number of target images not excluded at this time is equal to or smaller than the set maximum number of detections (see box Bo46 in FIG. 5). Determine whether. When the number of target images that are not excluded is larger than the maximum number of detections (step S440: No), the number of target images that exceed the maximum number of detections is determined so that the number of target images is equal to or less than the maximum number of detections. The items with the largest distance are selected and excluded in order (step S450). On the other hand, when the number of target images that are not excluded is equal to or smaller than the maximum number of detections (step S440: Yes), the process of step S450 is skipped.

  In step S460 (FIG. 10), the image search unit 210 (FIG. 1) determines whether all search stages have been selected. If there is still an unselected search stage (step S460: No), the unselected search stage having the smallest stage number is selected (step S420), and the search condition set for the selected search stage is used. The used image content use search is executed (steps S430 to S450). In this embodiment, in the condition determination in the second and subsequent search stages, only the feature amount distance calculated in the search stage is considered. However, in the condition determination in the second and subsequent search stages, the total distance between the feature amount distance calculated in the search stage and the feature amount distance calculated in the executed search stage is calculated, and the total distance is a threshold value. It may be determined whether or not the maximum feature amount total distance is less than or equal to. If it is determined in step S460 that all the search stages have been selected (step S460: Yes), the search execution process is completed. At this time, the target image that is not excluded becomes the detected image Di in the image search.

  When the search execution process (step S190 in FIG. 2) is completed, the window display control unit 211 (FIG. 1) controls the display processing unit 310 to display the search result window W3 on the display unit 150 (step S200). FIG. 11 is an explanatory diagram showing an example of the search result window W3. The search result window W3 is a window for displaying a list of target images (detected images Di) detected in the search execution process. That is, as shown in FIG. 11, the search result window W3 includes a detection image display area Ar62 in addition to the query image display area Ar61. In the detected image display area Ar62, the detected images Di are displayed side by side in a rank order (in ascending order of score) indicating the overall similarity to the query image. The score of each detected image Di is calculated based on the feature amount distance. In the present embodiment, the feature amount distance calculated in the final search stage of the image content use search is used as the score. This is because generally the most accurate signature method and metric type are often set for the final search stage. The score may be the sum or average of feature amount distances calculated in each search stage of image content use search, or may be the weighted sum or weighted average of feature amount distances calculated in each search stage. Also good.

  The search result window W3 (FIG. 11) includes a condition change button Bu61. When the condition change button Bu61 is selected by the user (step S210: Yes in FIG. 2), the search option window W2 (FIG. 5) is displayed again (step S150), and the search condition is reset and the search is re-executed (step S160 to S200) becomes possible.

  The detected image display area Ar62 of the search result window W3 (FIG. 11) includes query setting buttons Bu62 corresponding to the detected images Di. The query setting button Bu62 is a button for instructing to set the detected image Di as a new query image. When the query setting button Bu62 is selected by the user in the search result window W3 (step S220 in FIG. 2: Yes), the detected image Di corresponding to the selected query setting button Bu62 is a new query image. The resetting (step S230) is executed, and the search option window W2 is displayed again (step S150). In FIG. 12, when the query setting button Bu62 corresponding to the detected image Di (14) is selected in the search result window W3 (FIG. 11), the detected image Di (14) is set as a new query image. A search option window W2 is shown. The user can execute an image search using the detected image Di as a new query image without changing the search condition, or change the search condition.

  The search result window W3 (FIG. 11) includes a print designation area Ar63. In the search result window W3, for example, the user drags and drops the icon of the detection image Di to the print designation area Ar63, and displays the icon IC of the detection image Di in the print designation area Ar63, thereby detecting the detection image Di to be printed. Can be designated (step S240 in FIG. 2). In the example of FIG. 11, two detection images Di are designated as images to be printed. The print designation area Ar63 includes a print execution button Bu64 and a print designation release button Bu65. The user selects the icon IC of the image displayed in the print designation area Ar63 and selects the print designation release button Bu65 (or drags and drops the image icon IC onto the print designation release button Bu65). The print designation of the image can be canceled. In addition, the user can instruct the start of print execution of the detected image Di corresponding to the icon IC displayed in the print designation area Ar63 by selecting the print execution button Bu64. When an instruction to start printing is given, the print processing unit 320 (FIG. 1) generates print data based on the image data of the designated detection image Di, and controls the printer engine 160 to print the detection image Di. This is executed (step S250 in FIG. 2).

  As described above, in the image search / print processing by the printer 100 of this embodiment, the user can specify the search conditions for the image search in detail via the search option window W2 (FIG. 5). That is, the search option window W2 specifies a search stage definition area Ar46 for specifying a search condition relating to the feature of the image content in each of a plurality of search stages having a serial relationship, and each of the display states of the search stage definition area Ar46 The tag Ta41 corresponding to the search stage is included, and the user can easily specify the search condition of each search stage while switching the display state of the search stage defining area Ar46 via the tag Ta41. When a search condition is specified through the search option window W2, search conditions in each of a plurality of search stages are set according to the specification through the search option window W2, and a plurality of search stages are used using the set search conditions. The image search is sequentially executed. For this reason, in the image search / print processing by the printer 100 of the present embodiment, user convenience can be improved in image processing for performing image search.

  In the image search / print processing by the printer 100 of this embodiment, a plurality of search conditions (signatures) in which at least one of the speed and accuracy of the image search processing is different from each other in the search stage defining area Ar46 of the search option window W2. The search conditions can be specified by selecting from the choices of method, metric type, number of wavelet coefficients, etc., so the image content use search can be realized with the desired processing accuracy and processing speed. Convenience can be improved. For example, a search condition using a signature method (for example, a color histogram) that realizes a relatively high-speed and low-accuracy search process is set for the first search stage, and relatively slow for the second search stage. By setting a search condition using a signature method (for example, Haar wavelet) that realizes a high-precision search process, high-speed condition determination is performed for a large number of target images in the first search stage, and a high-precision The number of target images that are targets of the second search stage for performing the condition determination can be suppressed, and an image content use search with a good balance between processing accuracy and processing speed can be realized. Similarly, for example, for the first search stage, a search condition using a metric type (for example, metric M1) that realizes relatively high-speed and low-accuracy search processing is set, and for the second search stage, By setting a search condition using a metric type (for example, metric M3) that realizes a relatively low-speed and high-precision search process, it is possible to realize an image content use search with a good balance between the process accuracy and the process speed. it can. In these cases, by adjusting the maximum feature amount distance (that is, the threshold for condition determination) of the first search stage, the number of target images that are targets of the second search stage can be adjusted, It is possible to adjust the balance between the processing accuracy and the processing speed in the image content use search. In addition, each signature method and each metric type has a unique detection characteristic, so by performing a search using image content using multiple search stages that use different signature methods and metric types, the detection characteristic can be changed between the search stages. Complementation is performed, and the possibility that an image having a low similarity in human interpretation will be detected can be reduced. Further, in the image search / print processing by the printer 100 of the present embodiment, since the metadata use search and the image content use search can be combined, the user convenience in the image search can be further improved.

  In the image search / print processing by the printer 100 according to the present embodiment, the search option window W2 includes an interface (plus button Bu43 and minus button Bu44) for increasing or decreasing the number of search stages, so that the user can easily search for the search stage. The number can be specified, and the user convenience in the image content use search can be improved.

  Further, in the image search / print processing by the printer 100 of this embodiment, a query image is set as a reference for the image content use search, and a search condition regarding the similarity to the query image for the feature of the image content is set. It is possible to realize an image content use search that detects an image similar to the query image with respect to the characteristics of the predetermined image content. In the image search / print processing by the printer 100 of the present embodiment, the query image is specified by any one of the image file specification, the drawing, and the color specification in the initial window W1 (FIG. 3). Therefore, it is possible to improve user convenience in an image content use search for detecting an image similar to the query image. Further, in the image search / print processing by the printer 100 according to the present embodiment, one of the detected images Di detected by the image search can be set as a new query image. By setting the detected image Di as a new query image, the user can reach a desired search result more efficiently, and the user convenience in the image content use search can be further improved.

  Further, in the image search / print processing by the printer 100 of this embodiment, the search option window W2 includes the weight designation area Ar48 for designating the weight for each area of the image, so the similarity weighted for each area of the image. Search conditions can be set, and user convenience in image content use search can be further improved. Similarly, in the image search / print processing by the printer 100 of the present embodiment, the search option window W2 includes the box Bo44 for specifying the weight of each channel in the color space, so that the weight is assigned to each channel in the image color space. It is possible to set a search condition related to the similarity and further improve the user convenience in the image content use search.

  In the image search / print processing by the printer 100 of this embodiment, a detection image to be printed can be set from the detection images Di detected by the image search, and the detection image Di to be printed can be printed. . In the image search / print processing by the printer 100 of the present embodiment, the search result window W3 (FIG. 11) displays a detection image display area Ar62 for displaying a list of detection images Di, and a print specification for specifying the detection image Di to be printed. Since the detection image Di corresponding to the image (icon IC) displayed in the print designation area Ar63 is set as the detection image Di to be printed according to the user's predetermined operation, the user can easily perform the operation. The detected image Di to be printed can be specified, and the user convenience in the image search / print processing can be improved.

A-3. Feature distance calculation method:
Hereinafter, a specific method for calculating the feature amount distance by the similarity calculation unit 216 in the condition determination process (step S430) based on the feature amount distance of the search execution process (FIG. 10) described above will be described. The feature amount distance is an index representing the degree of similarity between the query image and the target image related to a predetermined feature of the image content.

A-3-1. Regarding the color histogram:
In the present embodiment, the metric type choices when the signature method is set to the color histogram in the image content usage search condition setting (step S180 in FIG. 2) are the standard histogram, the correlation histogram, the color moment, Four combined features are set (see FIG. 8). When the metric type is set to the standard histogram, the square D ² (Q, T) of the feature amount distance D (Q, T) is calculated by the following equation (6). In Expression (6), H _Qi is the pixel frequency of the i-th color bin of the query image, H _Ti is the pixel frequency of the i-th color bin of the target image, and N is the number of color bins.

When the metric type is set to the correlation histogram, the square D ² (Q, T) of the feature amount distance D (Q, T) is calculated by the following equation (7). In Expression (7), C _Qj is a cumulative value of the pixel frequencies H _Qi of the first to jth color bins of the query image (see Expression (1) above), and C _Tj is the first to j of the target image. This is the cumulative value of the pixel frequency H _Ti of the first color bin, and N is the number of color bins.

When the metric type is set to color moment, the feature amount distance D (Q, T) is calculated by the following equation (8). In Expression (8), D _RED (Q, T) is a feature amount distance in the R channel, and is calculated by the following Expression (9). In Equation (9), μ _Q and μ _T are the average pixel frequencies H _i in the color bin for the R channel of the query image and the target image, respectively (see Equation (2) above), and σ _Q and σ _T are respectively The square root (standard deviation) of the variance of the pixel frequency H _i in the color bin for the R channel of the query image and the target image (see equation (3) above), and γ _Q and γ _T are the R of the query image and the target image, respectively. It is the cube root of the skewness of the pixel frequency H _i in the color bin for the channel (see the above equation (4)), and w _μ , w _σ , and w _γ are weighting factors set by experiment. Further, D _GREEN (Q, T) and D _BLUE (Q, T) in equation (8) are feature amount distances in the G and B channels, respectively, and are calculated by the same equation as equation (9).

Even when the metric type is set to the combined feature, the feature amount distance D (Q, T) is calculated by the above equation (8). In this case, however, D _RED (Q, T) in equation (8) is calculated by the following equation (10). In the equation (10), F _Q and F _T are coupling indices for the R channel of the query image and the target image, respectively (see the above equation (5)), and w _R is a weighting factor set by experiment. D _GREEN (Q, T) and D _BLUE (Q, T) in Expression (8) are calculated by the same expression as Expression (10).

A-3-2. Regarding Haar wavelets:
In this embodiment, in the image content use search condition setting (step S180 in FIG. 2), the metric type options when the signature method is set to Haar wavelet are the high-speed and low-precision metric M1 and the medium-speed and medium-precision. Metric M2, low-speed and high-precision metric M3, and metric M4 as a real metric are set (see FIG. 8). As described above, the metrics M1, M2, and M3 differ only in the resolution of images when performing Haar wavelet decomposition, and the feature amount distance calculation method is the same. When the metric type is set to the metrics M1, M2, and M3, the feature amount distance D (Q, T) is calculated by the following equation (11). In Equation (11), k is the width and height of the target image of Haar wavelet decomposition, and Q (i, j) and T (i, j) are the coordinates (Haar wavelet decomposition results of the query image and target image, respectively) i, j) is the wavelet coefficient. Here, as the wavelet coefficient, a value obtained by performing quantization after a predetermined number of wavelet coefficients having smaller values are rounded down is used. Wavelet coefficient truncation is based on the wavelet coefficient number Cu specified in the box Bo49 of the search option window W2 (FIG. 9), leaving the wavelet coefficient of the wavelet coefficient number Cu having the largest value as it is, and other wavelet coefficients ( This is a process in which the value of the coefficient having a relatively small value is set to zero. In addition, the quantization of the wavelet coefficient in the present embodiment is a process of converting a positive value to +1 and a negative value to −1, and is defined by the following equation (12). In equation (11), w (i, j) is a weighting coefficient set by experiment. Further, in the equation (11), [Q (i , j) ≠ 1 T (i, j)] is a comparative value of the wavelet coefficients of wavelet coefficients and the target image of the query image, the following equation (13) It is prescribed by. That is, the feature amount distance D (Q, T) is a weighted sum of comparison values between the wavelet coefficient of the query image and the wavelet coefficient of the target image with respect to coordinates where the wavelet coefficient of the query image is not zero.

  The above equation (11) includes various improvements for speeding up the calculation processing of the feature amount distance D (Q, T). That is, in the above equation (11), the quantized wavelet coefficient is used, the scaling function coefficient (coefficient corresponding to the coordinates (0, 0)) is omitted, and the wavelet coefficient Q (i, Coordinates where j) is zero are excluded from the summation target, and only coordinates where Q (i, j) is not zero are subject to summation. Regarding the calculation of the feature distance D (Q, T) when the signature method is set to Haar wavelet, “Fast Multiresolution Image Querying” by CE Jacobs, A. Finkelstein and DH Salesin (Proceedings of 1995 ACM SIGGRAPH Conference , Los Angeles CA, USA, Aug. 9-11, pp. 277-286, 1995), and detailed description thereof is omitted here.

  In this embodiment, the real metric M4 as a metric type is a search condition used in an image search method using a search algorithm called Linear Approximating and Eliminating Search Algorithm (hereinafter also referred to as “LAESA” for short). LAESA is one of the pivot-based search algorithms. The pivot-based search algorithm calculates distances between a plurality of pivot points set in advance as pre-processing in order to reduce the distance calculation amount during the search processing. This is a search algorithm that detects points that are not likely to be satisfied and excludes them from the target of distance calculation. For the pivot-based search algorithm and LAESA, Maria Luisa Mico, Jose Oncina, “A new version of the Nearest-Neighbour Approximating and Eliminating Search Algorithm (AESA) with linear preprocessing time and memory requirements” (Pattern Recognition Letters vol.15, p.9-p.7 January 1994), Edgar Chavez, JL Marroquin, Ricardo Baeza-Yates, “Spaghettis: An Array Based Algorithm for Similarity Queries in Metric Spaces” (spire, pp. 38, String Processing and Information Retrieval Symposium & International Workshop on Groupware, 1999), detailed description is omitted here.

  When an image search method using a pivot-based search algorithm such as LAESA is employed, it is necessary to employ a symmetric method as a calculation method of the feature amount distance D (Q, T). The symmetric method is a feature amount distance calculation method in which the feature amount distance between the query image and the target image remains unchanged even when the relationship between the query image and the target image is reversed, and a method that satisfies the following equation (14) It is.

  When the above-described metric type is set to the metrics M1, M2, and M3, the feature amount distance D (Q, T) calculation method (the above formula (11)) has coordinates where Q (i, j) is zero. Since it is excluded from the summation target, it has no symmetry. Therefore, this calculation method cannot be used when the metric type is set to the real metric M4.

In this embodiment, the feature amount distance D (Q, T) when the metric type is set to the real metric M4 is calculated by a method defined by the following equation (15). In Equation (15), k is the width and height of the target image of Haar wavelet decomposition, and Q (i, j) and T (i, j) are the coordinates (Haar wavelet decomposition results of the query image and target image, respectively) i, j) is a wavelet coefficient (value after truncation and quantization), and w (i, j) is a weighting coefficient set by experiment. Further, in the equation (15), [Q (i , j) = 2 T (i, j)] is a comparative value of the wavelet coefficients of wavelet coefficients and the target image of the query image, the following equation (16) It is prescribed by.

  As shown in Expression (15), the feature amount distance D (Q, T) when the metric type is set to the real metric M4 is the coordinate for which the wavelet coefficient Q (i, j) of the query image is not zero. A weighted sum of the comparison values of the wavelet coefficients of the query image and the target image, and the sum of the weighting factors w (i, j) for coordinates where the wavelet coefficients T (i, j) of the target image are not zero; Is the sum of

  As can be seen by comparing the equation (15) with the above equation (11), in the equation (15), the weight coefficient w (i, j) corresponding to the coordinates where the wavelet coefficient T (i, j) of the target image is not zero. The sum is added as a correction term. The method of calculating the feature amount distance D (Q, T) defined by the equation (15) has symmetry as described below using the equation (17).

The uppermost stage of Expression (17) is the same as Expression (15). The second stage of Expression (17) is equivalent to the uppermost stage of Expression (17), as will be described below. That is, first, when Q (i, j) = T (i, j) (≠ 0), | Q (i, j) −T (i, j) | − | T (i, j) | = − | The equation T (i, j) | = −1 holds. Next, when Q (i, j) ≠ T (i, j) and T (i, j) = 0, | Q (i, j) −T (i, j) | − | T ( i, j) | = | Q (i, j) | = 1. Finally, when Q (i, j) ≠ T (i, j) and T (i, j) ≠ 0, | Q (i, j) −T (i, j) | − | T ( i, j) | = 2-1 = 1. Therefore, in all cases, [| Q (i, j) −T (i, j) | − | T (i, j) |] in the second stage of Expression (17) It is equal to [Q (i, j) = ₂ T (i, j)] in the upper stage.

  When T (i, j) ≠ 0, the equation | T (i, j) | = 1 holds, so the third stage of Expression (17) is equivalent to the second stage of Expression (17). is there. In the first term of the third stage of Expression (17), when T (i, j) = 0, | T (i, j) | = 0, so the condition of T (i, j) ≠ 0 Even if is removed, the sum of the first term is the same. Further, in the second term of the third stage of Expression (17), when Q (i, j) = 0, | Q (i, j) −T (i, j) | − | T (i, j ) | = 0, so that the sum of the second term is the same even if the condition of Q (i, j) ≠ 0 is removed. Therefore, the third level of equation (17) can be rewritten as the fourth level. Further, by combining the first term and the second term in the fourth stage of Expression (17), the fourth stage of Expression (17) can be rewritten as the fifth stage. The fifth stage of Expression (17) has symmetry because Expression (14) is established. Therefore, the equation (15) equivalent to the fifth stage of the equation (17) also has symmetry.

  As described above, the calculation method of the feature amount distance D (Q, T) defined by the above equation (15) has symmetry, and thus features when performing image search using a pivot-based search algorithm. It can be adopted as a method for calculating the quantity distance D (Q, T). The calculation method of the feature amount distance D (Q, T) defined by the above equation (15) is a comparison value between the wavelet coefficient Q (i, j) of the query image and the wavelet coefficient T (i, j) of the target image. Since the case where the wavelet coefficient Q (i, j) of the query image is zero is excluded in the calculation of the sum of the two (the second term on the right side of Expression (15)), the feature amount distance D (Q, T) The calculation speed of can be improved. Further, the first term (correction term) on the right side of the above equation (15) depends only on the target image and not on the query image, and thus can be calculated as a pre-processing of the image search processing. Therefore, the calculation method of the feature amount distance D (Q, T) in the above equation (15) realizes an increase in the processing speed of the image search processing using a pivot-based search algorithm such as LAESA and a reduction in the processing time. can do.

B. Second embodiment:
FIG. 13 is an explanatory diagram schematically showing a configuration of a printer 100a as an image processing apparatus in the second embodiment. In the printer 100 a of the second embodiment, the image search unit 210 includes an allowable time setting unit 218 and a minimum image number setting unit 219, and the option table CT is stored in the internal memory 120. This is different from the first embodiment shown in FIG. Other configurations of the printer 100a are the same as those in the first embodiment.

  FIG. 14 is a flowchart showing the flow of image search / print processing in the second embodiment. In the image search / print processing (FIG. 2) of the first embodiment, the number of search stages and the search conditions in each search stage are set according to the user's specification. In the image search / print processing of the second embodiment, search is performed. The number of stages and search conditions at each search stage are automatically set.

  When the image search / print process (FIG. 14) is started, a query image is first set (step S120). The query image setting process is executed in the same manner as in the first embodiment. Next, search condition setting processing is executed (step S132). FIG. 15 is a flowchart showing the flow of search condition setting processing. The search condition setting process is a process for automatically setting the number of search stages in the image search and the search conditions in each search stage.

  In the search condition setting process of this embodiment, automatic setting is performed for two of the metric type and the maximum feature amount distance among the elements (see FIG. 9) that define the search conditions of the search stage. As described above, the metric type defines an index value that represents the degree of similarity of the image content feature with the query image and a method for calculating the index value. The maximum feature amount distance is a threshold value in condition determination based on the feature amount distance. For other elements that define the search conditions of the search stage, preset values are adopted. In this embodiment, Haar wavelet is adopted as a signature method, RGB is adopted as a color space, default values are adopted for the number of wavelet coefficients and the maximum number of detections, and color channel weights and division area weights are set to default values (all weights). Is set to 1.0). In this embodiment, it is assumed that the metadata use search is not executed.

  Metric type options that can be selected in the search condition setting process are set in advance and defined in the option table CT. FIG. 16 is an explanatory diagram illustrating an example of the option table CT. In the option table CT of this embodiment, three types of metrics, that is, a high-speed / low-accuracy metric M1, a medium / medium-accuracy metric M2, and a low-speed / high-accuracy metric M3 are set. In this embodiment, the processing speed of each metric type is represented by processing time (seconds) per 1,000 images. This processing speed is an experimental value or a theoretical value when processing is performed by the printer 100a.

  FIG. 17 is an explanatory diagram specifically showing how search stages and search conditions are set in the search condition setting process. The search condition setting process will be described below with reference to the flowchart in FIG. 15 and the explanatory diagram in FIG.

  In step S520 of the search condition setting process (FIG. 15), the allowable time setting unit 218 (FIG. 13) sets the allowable required time Tmax according to the user's specification via a user interface (not shown). The allowable required time Tmax is the longest time allowed as the time required for image search through all search stages. That is, in the search condition setting process, the number of search stages and the search conditions in each search stage are set within a range where image search through all search stages is completed within the allowable required time Tmax. Note that a default value may be used for the allowable required time Tmax.

  In step S530 (FIG. 15), the minimum image number setting unit 219 (FIG. 13) sets the minimum detected image number NDmin in accordance with a user designation via a user interface (not shown). The minimum number of detected images NDmin is the minimum value of the number of detected images Di detected in the image search through all the search stages. That is, in the search condition setting process, the number of search stages and the search conditions in each search stage are set in a range where the number of detected images Di detected in the image search through all the search stages is equal to or greater than the minimum detected image number NDmin. Is done. The minimum number of detected images NDmin is set based on, for example, the number of detected images Di that can be displayed within one page in the search result window W3 (FIG. 11). A default value may be used for the minimum number of detected images NDmin.

  In step S540 (FIG. 15), the search condition setting unit 213 (FIG. 13) selects the most accurate metric type as the metric type for the main stage Sm. Here, the main stage Sm is a search stage executed last. As shown in FIG. 16, in this embodiment, the metric type with the best accuracy is the metric M3, and therefore the metric M3 is set as the metric type for the main stage Sm. In case 1 shown in FIG. 17, the metric M3 is selected as the metric type for the main stage Sm.

  In step S550 (FIG. 15), the search condition setting unit 213 (FIG. 13) calculates the required search time Trm of the main stage Sm, and determines whether the required search time Trm is less than or equal to the allowable required time Tmax. If it is determined that the required search time Trm is less than or equal to the allowable required time Tmax, it is possible to complete the search process within the allowable required time Tmax with only the main stage Sm that employs the most accurate metric type. Therefore, a search condition composed only of the main stage Sm adopting the metric type having the best accuracy is set. For example, the required search time Trm by the main stage Sm adopting the metric M3 is 0.8 seconds (see FIG. 16). In the description of this embodiment, time is expressed as time per 1,000 images. In case 1 shown in FIG. 17, since the allowable required time Tmax is set to 0.9 seconds, it is determined that the required search time Trm is less than or equal to the allowable required time Tmax. Accordingly, in case 1, a search condition constituted only by the main stage Sm adopting the metric M3 is set. In this case, the search condition setting unit 213 determines the maximum feature amount distance (threshold value) of the main stage Sm based on the maximum number of detections (step S640), and the search condition setting process is completed.

  If it is determined in step S550 (FIG. 15) that the required search time Trm is greater than the allowable required time Tmax, only the main stage Sm adopting the metric type having the best accuracy performs the search process within the allowable required time Tmax. Cannot be completed. In this case, the search condition setting unit 213 (FIG. 13) determines whether or not there is a metric type option faster than the metric type set in the main stage Sm (step S560). If it is determined in step S560 that there is no faster metric type option, it is impossible to complete the image search within the allowable required time Tmax, so the metric type option is changed and set. (Step S670), the processing after step S540 is executed again.

  When it is determined in step S560 (FIG. 15) that there is a faster metric type option, the search condition setting unit 213 (FIG. 13) selects the metric type with the highest speed among the options for the previous stage Sp. Is selected as the metric type (step S570). In case 2 shown in FIG. 17, since the allowable required time Tmax is set to 0.3 seconds, it is determined in step S550 that the required search time Trm (0.8 seconds) is larger than the allowable required time Tmax ( (Refer to the first stage of Case 2). In this case, the metric M1 (see FIG. 16), which is the metric type with the highest speed, is set as the metric type for the previous stage Sp (see the second stage in Case 2).

  In step S580 (FIG. 15), the search condition setting unit 213 (FIG. 13) calculates the required search time Trp of the preceding stage Sp, and determines whether the required search time Trp is less than or equal to the allowable required time Tmax. If it is determined that the required search time Trp of the preceding stage Sp is larger than the allowable required time Tmax, it is impossible to complete the image search within the allowable required time Tmax even if the metric type with the highest speed is adopted. Therefore, the metric type option change setting is performed (step S670), and the processing from step S540 onward is executed again.

  If it is determined in step S580 (FIG. 15) that the required search time Trp of the preceding stage Sp is less than or equal to the allowable required time Tmax, the search condition setting unit 213 (FIG. 13) allows the allowable input image of the main stage Sm. The number NImax is calculated (step S590). The allowable input image number NImax is the maximum number allowed as the number of target images that are the target of the main stage Sm. The allowable input image number NImax is calculated by dividing the difference between the allowable required time Tmax and the required search time Trp by the metric type speed of the main stage Sm. In case 2 shown in FIG. 17 (see the second stage of case 2), the metric M3 is set in the main stage Sm, and the metric M1 is set in the preceding stage Sp. ) And the required search time Trp (0.1 seconds) is divided by the speed of the metric M3 (0.8 seconds / 1,000 images), and the allowable input image number NImax is 250 images. Is calculated.

  In step S600 (FIG. 15), the search condition setting unit 213 (FIG. 13) determines the maximum feature amount distance (threshold value) of the preceding stage Sp based on the allowable input image number NImax. The maximum feature amount distance of the preceding stage Sp is determined so that the number of detected images Di in the preceding stage Sp is equal to or less than the allowable input image number NImax.

  In step S610 (FIG. 15), the search condition setting unit 213 (FIG. 13) determines whether or not the number of detected images ND is greater than or equal to the minimum number of detected images NDmin. The detected image number ND is the number of detected images Di detected in the image search through all the search stages (the main stage Sm and the previous stage Sp). If it is determined in step S610 that the detected image number ND is equal to or greater than the minimum detected image number NDmin, the search condition setting unit 213 determines the maximum feature amount distance (threshold value) of the main stage Sm based on the maximum detected number. (Step S640), and the search condition setting process is completed.

  If it is determined in step S610 (FIG. 15) that the detected image number ND is smaller than the minimum detected image number NDmin, the search condition setting unit 213 (FIG. 13) has not been selected as the metric type for the main stage Sm. It is determined whether there is an option (step S620). If it is determined in step S620 that there are no unselected options, metric type option change setting is performed (step S670), and the processes in and after step S540 are executed again.

  On the other hand, if it is determined in step S620 (FIG. 15) that there are unselected options, the search condition setting unit 213 selects the metric type having the best accuracy among the unselected options for the main stage Sm. The metric type is selected (step S630). At this time, the previous stage Sp is canceled. In case 3 shown in FIG. 17, when the metric M3 is set in the main stage Sm and the metric M1 is set in the previous stage Sp, it is determined that the detected image number ND is smaller than the minimum detected image number NDmin. (Refer to the second stage of Case 3). In this case, the metric M2 (see FIG. 16), which is the most accurate metric type among the unselected options, is selected as the metric type for the main stage Sm (see the third stage of case 3).

  After the metric type for the main stage Sm is reselected in step S630 (FIG. 15), the processing from step S550 is executed again. In case 3 shown in FIG. 17 (see the third stage of case 3), in the determination in step S550 after the metric M2 is set in the main stage Sm, the required search time Trm (0.5 seconds) of the main stage Sm. Is determined to be greater than the allowable required time Tmax (0.2 seconds). In this case, the metric M1 with the highest speed is selected as the metric type for the preceding stage Sp (step S570), the allowable input image number NImax is calculated (step S590), and the maximum feature amount distance (threshold value) of the preceding stage Sp. Is determined (step S600), and it is determined whether or not the number of detected images ND is greater than or equal to the minimum number of detected images NDmin (step S610). If it is determined in step S610 that the number of detected images ND is equal to or greater than the minimum number of detected images NDmin (see the fourth stage of case 4), the maximum feature amount distance (threshold) of the main stage Sm is determined (step S610). S640), the search condition setting process is completed.

  When the search condition setting process (step S132 in FIG. 14) is completed, the process after the search execution process (step S190 in FIG. 14) using the set search condition is performed as in the first embodiment shown in FIG. Executed.

  As described above, in the image search / print process according to the second embodiment, the allowable required time Tmax for image search is set. Based on the allowable required time Tmax, the number of search stages for image search and the search conditions in each search stage are set. Is set, and image search of the search stage using the set search condition is sequentially executed. Therefore, in the image search / print process of the second embodiment, it is possible to automatically set the number of search stages and search conditions in the image search, thereby improving user convenience.

  That is, in the image search / print processing of the second embodiment, a query image is set as a reference for image search, and an index value (feature value) representing the similarity of the image content feature to the query image for each search stage. The search condition for specifying the distance), the index value calculation method (metrics M1, M2, M3) and the threshold value (maximum feature amount distance) for the index value is set, and the calculation method set in each search stage is used. An index value is calculated, and an image is detected by determination using a threshold value. Therefore, in the image search / print processing of the second embodiment, the number of search stages in image search, the index value indicating similarity, the index value calculation method, and the search condition for specifying the threshold value for the index value are automatically set. User convenience, and user convenience can be improved.

  In the image search / print process of the second embodiment, one of a plurality of preset calculation methods (metrics M1, M2, M3) in which at least one of processing speed and processing accuracy is different from each other is selected. Since the search condition is set, the optimum number of search stages and the search condition considering the balance between the processing speed and the processing accuracy are automatically set, and the user convenience can be further improved. That is, in the image search / print processing of the second embodiment, the search condition is selected so that a calculation method with better processing accuracy is selected within a range where the image search through all the search stages is completed within the allowable required time Tmax. Therefore, the optimum number of search stages and search conditions are automatically set in consideration of the balance between processing speed and processing accuracy.

  In the image search / print process of the second embodiment, the minimum number of detected images NDmin is set, and the range in which the number of detected images Di detected in the image search through all the search stages is equal to or greater than the minimum number of detected images NDmin. Therefore, since the number of search stages and the search conditions at each search stage are set so that more search conditions for realizing an image search with good processing accuracy are adopted, the number of detected images Di is too small. The optimum number of search stages and search conditions are automatically set in consideration of the balance between the processing speed and the processing accuracy within the range where there is no such problem.

  Further, in the image search / print process of the second embodiment, the detected image Di detected in the image search using the automatically set search condition can be printed.

C. Variation:
The present invention is not limited to the above-described examples and embodiments, and can be implemented in various modes without departing from the gist thereof. For example, the following modifications are possible.

C1. Modification 1:
In each of the above embodiments, the image search / print processing is executed using the image (image data (image file)) stored in the memory card MC as the target image. However, the target image in the image search / print processing is arbitrarily selected. It can be set. For example, an image stored in a predetermined area of the internal memory 120 (FIG. 1) may be set as a target image, or in a storage area of another device connected to the printer 100 via a cable or a network. The stored image may be set as the target image.

C2. Modification 2:
The contents and layout of the initial window W1, the search option window W2, and the search result window W3 in the above embodiments are merely examples, and various modifications can be made. These windows can be modified as appropriate according to processing that can be executed by the printer 100. For example, in each of the above embodiments, the printer 100 can specify a query image by three methods (image file designation, drawing, and color selection). However, in any of the three methods, a query image can be designated. The query image may not be specified, or the query image may be specified by a method other than the three methods (for example, a method of selecting a preset template image). The initial window W1 (FIG. 3) is appropriately modified according to the query image designation method possible in the printer 100.

  Further, the options (camera manufacturers, camera models, file sizes, etc., see FIG. 7) in each element that define the search conditions for the metadata use search described in the above embodiments are merely examples, and can be variously modified. . The search option window W2 (FIG. 5) is appropriately modified according to the deformation of these elements and options. Further, the printer 100 does not need to be able to execute the metadata use search, and the search option window W2 does not have to include the metadata condition specifying area Ar43 for specifying the search condition for the metadata use search.

  In addition, the elements that define the search conditions for the image content use search described in the above embodiments (signature method, metric type, color space, etc., see FIGS. 5 and 9), and options in each element (color histogram, Haar wavelet, etc.) FIG. 8) is merely an example, and various modifications can be made. The search option window W2 (FIG. 5) is appropriately modified according to the deformation of these elements and options. For example, color channel weights and divided region weights do not need to be included as elements that define the search conditions. In addition, other options such as a color space, a SHA hash sum, and a wavelet coefficient using a wavelet base other than the Haar wavelet may be set as a signature method option that is one of the elements that define the search condition. Further, as a metric type option when the signature method is set to Haar wavelet, a plurality of feature value distance calculation method options having different wavelet coefficient numbers (see box Bo49 in FIG. 9) may be set.

C3. Modification 3:
The method for calculating the feature amount distance as an index value representing the degree of similarity between the query image and the target image in each of the above embodiments is merely an example, and various modifications can be made. For example, when the signature method is set to Haar wavelet and the metric type is set to metric M1, the calculation formula for the feature amount distance D (Q, T) (the expression (11) uses a query for speeding up the processing). The coordinates where the wavelet coefficient Q (i, j) of the image is zero are excluded from the object of summation, but the coordinates where Q (i, j) is zero may be the object of summation. 11), the wavelet coefficients Q (i, j) and T (i, j) of the query image and the target image are quantized, but the coefficients before being quantized are used. The same applies when the metric type is set to real metric M4.

C4. Modification 4:
In the search condition setting process (FIG. 15) in the second embodiment, the number of search stages and the metric type at each search stage are automatically set. In the search condition setting process, the signature method at each search stage is set. And a combination of metric type may be automatically set. That is, combinations of a plurality of signature methods and metric types that differ in at least one of processing speed and processing accuracy are preset, and a specific combination is selected from a plurality of options based on the determination using the allowable required time Tmax. By selecting, the signature method and the metric type may be automatically set. Further, in the search condition setting process, the necessity of executing the metadata use search and the search condition of the metadata use search may be automatically set. For example, the matching item rate (ratio of the number of condition matching items to the number of items for which conditions are set (camera maker and file size)) as a threshold for condition determination in metadata search and a plurality of keywords can be used The matching keyword rate (ratio of the number of matching keywords to the set number of keywords) as a threshold for condition determination in the metadata use search may be automatically set. These are automatically set based on the allowable input image number NImax, for example, when the allowable input image number NImax of the main stage Sm is calculated (step S590 in FIG. 15). In this case, the match item rate and the match keyword rate are used as index values representing the similarity between the query image and the target image.

  In the search condition setting process (FIG. 15) of the second embodiment, the search condition is automatically configured to include one (main stage Sm) or two (main stage Sm and previous stage Sp) search stages. Although it is set, the search condition may be automatically set so as to include three or more search stages. In the search condition setting process of the second embodiment, the setting of the minimum detected image number NDmin (step S530) and the determination using the minimum detected image number NDmin (step S610) can be omitted.

C5. Modification 5:
The configuration of the printer 100 as the image processing apparatus in each of the above embodiments is merely an example, and the configuration of the printer 100 can be variously changed. In each of the above-described embodiments, the image search / print process by the printer 100 as the image processing apparatus has been described. It may be executed by the image processing apparatus. The printer 100 is not limited to an ink jet printer, and may be another type of printer such as a laser printer or a sublimation printer.

  In each of the above embodiments, a part of the configuration realized by hardware may be replaced with software, and conversely, a part of the configuration realized by software may be replaced by hardware. .

  In addition, when part or all of the functions of the present invention are realized by software, the software (computer program) can be provided in a form stored in a computer-readable recording medium. In the present invention, the “computer-readable recording medium” is not limited to a portable recording medium such as a flexible disk or a CD-ROM, but an internal storage device in a computer such as various RAMs and ROMs, a hard disk, etc. It also includes an external storage device fixed to the computer.

100 ... Printer 110 ... CPU
DESCRIPTION OF SYMBOLS 120 ... Internal memory 140 ... Operation part 150 ... Display part 160 ... Printer engine 170 ... Card interface 172 ... Card slot 200 ... Image processing part 210 ... Image search part 211 ... Window display control part 212 ... Query image setting part 213 ... Search condition Setting unit 216 ... Similarity calculation unit 217 ... Print image setting unit 218 ... Allowable time setting unit 219 ... Minimum image number setting unit 310 ... Display processing unit 320 ... Print processing unit

Claims (17)

An image processing apparatus for performing image search,
A stage-specific condition designation area for designating a search condition relating to the characteristics of image contents in each of a plurality of search stages having a serial relationship, and a tag corresponding to each of the plurality of search stages, the stage-specific condition designation area A condition specification window display for displaying a condition specification window including a plurality of tags for specifying the display state of
A search condition setting unit that sets the search condition in each of the plurality of search stages according to the specification through the condition specifying window;
An image search unit that sequentially executes image search of the plurality of search stages using the set search conditions. The image processing apparatus according to claim 1,
The stage-specific condition designation area is an area for designating one of the plurality of search conditions that realizes an image search in which at least one of processing speed and processing accuracy is different as the search condition to be adopted. Processing equipment. The image processing apparatus according to claim 1 or 2, wherein
The condition designating window display unit displays the condition designating window including a stage number designating area for designating the number of the search stages,
The image processing apparatus, wherein the search condition setting unit sets the search condition in each of a number of the search stages specified via the condition specifying window. The image processing apparatus according to any one of claims 1 to 3, further comprising:
A query image setting unit for setting a query image as a reference for image search;
The said search condition setting part is an image processing apparatus which sets the said search condition regarding the similarity with the said query image about the said characteristic. The image processing apparatus according to claim 4,
The query image setting unit sets an image specified by any one of an image file designation, a drawing, and a color designation as the query image. The image processing apparatus according to claim 4 or 5, wherein
The said query image setting part is an image processing apparatus which sets one of the images detected by the image search as a new said query image. An image processing apparatus according to any one of claims 4 to 6,
The said search condition setting part is an image processing apparatus which sets the said search condition regarding the said similarity weighted for every area | region of the image. An image processing apparatus according to any one of claims 4 to 7,
The image processing apparatus, wherein the search condition setting unit sets the search condition related to the similarity weighted for each channel of an image color space. An image processing apparatus according to any one of claims 1 to 8,
The image search unit is an image processing device that performs an image search of the subsequent search stage for an image detected in the image search of the previous search stage in the serial relationship. An image processing apparatus according to any one of claims 1 to 9, wherein
The search condition setting unit sets a search condition related to the attribute of the image in addition to the search condition related to the feature of the image content,
The image search unit is an image processing device that performs an image search using a search condition relating to an attribute of the image and a search condition relating to a feature of the image content. An image processing apparatus according to any one of claims 1 to 10,
The condition designation window display unit displays the condition designation window in which the stage-specific condition designation areas corresponding to only one search stage are displayed at the same time according to designation via the tag. An image processing apparatus according to any one of claims 1 to 11,
The image processing apparatus, wherein the feature is at least one of a feature representing a color distribution in an image and a feature calculated by wavelet decomposition of the image. The image processing apparatus according to any one of claims 1 to 12, further comprising:
A print image setting unit for setting a detection image to be printed from detection images detected by image search;
An image processing apparatus comprising: a printing unit that prints a detection image to be printed. The image processing apparatus according to claim 13, further comprising:
A search result window display unit for displaying a search result window including a detection image display area for displaying a list of detection images and a print designation area for designating a detection image to be printed;
The print image setting unit, in the search result window, sets a detection image corresponding to an image displayed in the print designation area as a detection image to be printed in response to a user's predetermined operation. An image processing method for performing image search,
(A) a stage-specific condition designation area for designating a search condition relating to the feature of the image content in each of a plurality of search stages having a serial relationship, and a tag corresponding to each of the plurality of search stages, A step of displaying a condition designation window including a plurality of tags for designating a display state of the condition designation area;
(B) setting the search condition in each of the plurality of search stages according to the specification through the condition specifying window;
(C) An image processing method comprising: sequentially executing image searches of the plurality of search stages using the set search conditions. A computer program for image processing that performs image search,
A stage-specific condition designation area for designating a search condition relating to the characteristics of image contents in each of a plurality of search stages having a serial relationship, and a tag corresponding to each of the plurality of search stages, the stage-specific condition designation area A condition specification window display function for displaying a condition specification window including a plurality of tags for specifying the display state of
A search condition setting function for setting the search condition in each of the plurality of search stages according to the specification through the condition specifying window;
A computer program that causes a computer to realize an image search function that sequentially executes image search of the plurality of search stages using the set search conditions. A printing device,
A stage-specific condition designation area for designating a search condition relating to the characteristics of image contents in each of a plurality of search stages having a serial relationship, and a tag corresponding to each of the plurality of search stages, the stage-specific condition designation area A condition specification window display for displaying a condition specification window including a plurality of tags for specifying the display state of
A search condition setting unit that sets the search condition in each of the plurality of search stages according to the specification through the condition specifying window;
An image search unit that sequentially executes image search of the plurality of search stages using the set search conditions;
And a printing unit that prints an image detected by the image search.

Images